PART III - TRANSCATHETER TRICUSPID VALVE INTERVENTIONS
TABLE OF CONTENTS
BOOKSHOP
Updated on August 27, 2021
PART III

Transcatheter tricuspid valve interventions

Jean-Michel Juliard, Giulio Russo1, 2, Alec Vahanian, Azeem Latib, Eric Brochet, Phalla Ou, Dominique Himbert, Maurizio Taramasso1, Francesco Maisano1
1.University Heart Centre, University Hospital, Zurich, Switzerland
2.Fondazione Policlinico Universitario A. Gemelli IRCCS, Università Cattolica del Sacro Cuore, Rome, Italy

Summary

Functional (or secondary) tricuspid regurgitation (TR) is the most frequent tricuspid valve disease and correlates with an increased risk of cardiovascular events and death at long-term follow-up. Functional TR is due to tricuspid annulus dilatation without an anatomical lesion of the leaflets. Most often, it is observed in patients with left-heart valvular disease (with or without previous surgical correction) and pulmonary artery hypertension. It may also be due to right ventricular failure of any cause, usually associated with permanent atrial fibrillation. Transthoracic and transesophageal echocardiography (2D and 3D) are mandatory for TR assessment: severity, measurement of the tricuspid annulus, analysis of the mechanisms and valve leaflet anatomy. Alongside this cardiac computed tomography is used routinely to plan the procedures while intracardiac echocardiography is a complementary tool used in a minority of transcatheter procedures.

In patients with previous tricuspid surgery, transcatheter valve-in-valve or valve-in-ring procedures can be performed using a transcatheter aortic prosthesis and a transfemoral or transjugular approach. In patients with native disease, innovative percutaneous techniques mimic surgical techniques and aim at correcting valvular disease through different mechanism such as leaflet grasping, annuloplasty (either direct or indirect), caval valve implantation or valve replacement.

Introduction

Tricuspid valve disease

Tricuspid valve disease can be organic (primary) or functional (secondary).

Primary TR is rare, and associated with organic lesions of the leaflets and chordae, including numerous aetiologies: congenital, infective endocarditis, neoplastic, rheumatic, endomyocardial fibrosis, traumatic or iatrogenic causes (following percutaneous lead implantation or right ventricular biopsy). Under these circumstances, surgical repair and replacement or percutaneous replacement usually play a primary role.

Functional TR is the most frequent tricuspid valve (TV) disease encountered in Western countries and its prevalence is increasing [1]. Functional TR is mainly due to a combination of annular dilatation and leaflet tethering, causing a lack of leaflets coaptation and various degrees of TR severity. Most often, it develops progressively in patients with left-sided heart valve disease or pulmonary artery hypertension, in the absence of organic lesions of the leaflet. Functional TR may also be secondary to right ventricular (RV) failure of any cause. It is also often associated with permanent atrial fibrillation. The presence of functional TR, either isolated or in combination with left heart disease, is associated with unfavourable natural history. Although functional TR responds to medical therapy, in the later stages tricuspid repair and replacement might be needed. However, tricuspid surgery in this context is often carried out with high morbidity and mortality, since often the patients are referred to surgery late, when RV failure is advanced. This evidence has supported early interventions (combined tricuspid repair at the time of the left-sided disease treatment) and is the basis for the research efforts to identify innovative percutaneous techniques.

In addition, tricuspid recurrent disease occurring in patients with previous surgical correction, either with a ring annulus or with a bioprosthetic valve, could be treated with valve-in-valve or valve-in-ring approaches in some carefully selected patients.

Epidemiology of TR

Moderate-to-severe TR is present in 1.6 million US individuals and only a few (<0.5%) of this population currently undergo surgical tricuspid repair or replacement [2]. In the European Heart Survey on valvular heart disease published in 2003, 72% had native valve disease (only 1% with isolated tricuspid valve disease, but 20% with multiple valve disease, especially TR associated with mitral disease) [3].

The difference between the high prevalence of the disease and the incidence of interventional therapy is due to two factors: the perception that TR has minimal prognostic impact, and the risk of treating these patients. The prevalence of moderate to severe TR is particularly frequent in patients with mitral valve disease: around one third of patients with either degenerative or functional mitral regurgitation or rheumatic mitral stenosis have significant TR at the time of the mitral surgery [4, 5]. After isolated mitral valve repair, significant residual TR is observed in up to 40% of patients [6, 7]. In patients undergoing concomitant TR repair at the time of mitral surgery, persistent severe TR is still present in 15%. Predictors for residual regurgitation after surgical repair have been identified: higher preoperative TR severity, higher pulmonary artery pressures, mitral replacement rather than repair, worse left ventricular dysfunction, presence of pacemaker leads through the valve area [8]. Surgical ring repair seems to be generally more durable and with less residual TR as compared to suture repairs [9].

Current management of TR

Tricuspid regurgitation does not have the same immediate impact on short-term mortality in comparison to other heart valve diseases. In the absence of pulmonary hypertension, TR is well tolerated on long-term follow-up. Diuretics and vasodilators are very effective in reducing right heart failure symptoms in this setting, until end-stage right heart failure develops. In this situation, cardiac surgeons are reluctant to operate for isolated TR in patients with previous left-sided heart valve surgery or poor RV function, since the risk is often too high [10]. Rates of in-hospital mortality for isolated tricuspid valve surgery are not consistent in the literature, ranging from 2 to 9.8% [11]. Older age, severe hepatic and/or renal impairment, anaemia and albumin before intervention are independent determinants of hospital mortality [12]. On the other hand, the type of procedure (replacement versus repair), cause of TR (primary isolated versus secondary) and the presence of leaflet abnormality are not predictive of mortality [12].

Recent evidence has demonstrated that untreated TR might be associated with worse prognosis [13, 14]. Persistent functional TR is associated with excess mortality on long-term follow-up and also with recurrent hospitalisations for right heart failure leading to a progressive reduction of functional capacity and eventually to a very severe disability despite high dosage of diuretics [15, 16]. Moreover, recurrent hospitalisations represent a substantial cost for the healthcare system.

Since significant TR persists after left-sided heart valve surgery, a lower threshold for TR treatment at the time of left-sided intervention is today recommended by international guidelines. Concomitant tricuspid valve annuloplasty with mitral valve surgery, based on tricuspid annular dilatation and irrespective of the grade of regurgitation, leads to a significant improvement in functional status and prevents the development of clinically-relevant regurgitation in the long-term [17].

Guideline recommendations for tricuspid surgery

Guideline recommendations for surgery are currently based upon expert opinion in the absence of evidence-based trial data ( Table 1). The ideal timing of surgery for TV disease remains controversial, mostly due to the limited data available and their heterogeneous nature. When technically feasible, valve repair is always preferable to valve replacement. Regarding the opportunity of TR repair in patients with a planned left-sided heart valve surgery and functional TR, American and European guidelines are quite comparable, with the same threshold for a dilated tricuspid annulus of 40 mm or >21 mm/m2, regardless of the severity of the TR [18, 19].

The strict adherence to these guidelines could contribute to reducing the number of patients with severe, symptomatic TR, especially after left-sided heart valve surgery and in those at high risk for redo surgery.

Patients who may be suitable for percutaneous approach

Percutaneous procedures may be an attractive alternative to surgery for patients with symptomatic functional TR or severe deterioration of the TV after surgical repair and deemed to be high-risk surgical candidates.

Further to the indication for patients with a history of cardiac surgery, a growing population of patients with mitral regurgitation is currently being treated with percutaneous techniques. Percutaneous treatment of functional mitral regurgitation with the MitraClip system (Abbott Vascular, Santa Clara, CA, USA) in high-risk surgical patients is now well established and this population is growing. Moderate to severe TR at baseline is frequent in these patients (47 out of 146 consecutive patients in the GRASP registry) and has a negative impact on the 30-day and 12-month outcome on the combined risk of death and rehospitalisation for heart failure [20]. In this specific population, a simultaneous or staged percutaneous treatment of TR could be a safe option to improve outcomes.

Anatomical considerations for transcatheter tricuspid interventions

In-depth knowledge of the complex surgical anatomy of the TV is a mandatory prerequisite to understand the challenges, which are encountered in developing percutaneous tricuspid therapy.

The normal TV apparatus is composed of three leaflets (septal, posterior and anterior), chordae tendinae, and two or three well developed papillary muscles. The posterior and anterior leaflets could be considered scallops of a single mural leaflet, similar to the posterior leaflet of the mitral valve. Consequently, variable distribution of the scallops is possible and more than three leaflets can be identified in many patients, depending on the pattern of the subdivision of the mural leaflet. Papillary muscle anatomy is also variable, with a septal and an anterior papillary muscle. The septal leaflet has direct chordal attachments to the septum without the presence of a developed papillary muscle in most cases. Several structures of major surgical importance surround the TV, such as the right coronary artery (which is very often in close contiguity with the parietal portion of the tricuspid annulus), the atrioventricular (AV) node, the bundle of His and the coronary sinus ( Figure 1). The non-planar and non-circular structure of the tricuspid annulus must be taken into account when considering tricuspid repair technologies. The normal shape of the tricuspid annulus is semi-lunar and it is larger compared to the mitral one. Under physiological conditions, the tricuspid annulus has a non-planar three-dimensional saddle-shaped configuration. Such a characteristic is lost in cases of functional TR, since the annulus becomes larger and flat. In the presence of functional TR, as a consequence of RV enlargement and dysfunction, the tricuspid annulus becomes more circular as it dilates in its anterior-posterior aspect. The consequence of severe functional TR is further RV dilatation with progressive papillary muscle displacement and consequent leaflet tethering. Focus box below summarises the major anatomical challenges to be overcome in the development of transcatheter tricuspid valve therapies.

FOCUS BOX 1Anatomical challenges for transcatheter tricuspid valve therapies.
    • Large tricuspid annulus dimensions with a non-planar and elliptical shape
    • Absence of calcifications
    • RV morphology (trabeculae, muscle bands, thin apical wall)
    • Proximity/contiguity of other structures (coronary sinus, AV node and bundle of His, venae cavae, right coronary artery)
    • Angulation of the annulus in relation to the superior vena cava and inferior vena cava

Imaging of the tricuspid valve: patient selection and procedural guidance

Pre-procedural echocardiography

Echocardiography is the essential imaging technique in the assessment of patients with TR. Two-dimensional (2D) transthoracic echocardiography (TTE) is the imaging technique of choice to assess the mechanism and severity of TR. Echocardiography is also of help to define TV morphology and RV anatomy and function. In addition, three-dimensional (3D) TTE is increasingly recognised as an important adjunct to 2D TTE for a more comprehensive assessment of the complex morphology of the TV. 2D and 3D transoesophageal echocardiography (TOE) as well as intracardiac echocardiography (ICE) are essential for procedural guidance.

The main echographic views used to assess leaflet anatomy of the TV with 2D TTE are the parasternal long-axis view of RV inflow, the parasternal short-axis view at the level of the aortic valve, and the apical four-chamber and subcostal views. Due to the complex morphology of the TV, only two leaflets can generally be visualised at a time, with some uncertainty regarding their precise identification, according to the various cutting planes. Commissures are difficult to assess with 2D techniques, unless a short-axis view of the TV can be obtained (low parasternal and subcostal short-axis view) [21, 22, 23]. Conversely, real-time 3D TTE, when available, may provide an “en face” view of the TV which allows visualisation of the three leaflets simultaneously, precise identification of the commissures and assessment of leaflet coaptation [21].

Using TTE, the tricuspid annular diameter is generally measured from the apical four-chamber view, at end-diastole. The normal annulus diameter in adults is 28±5 mm and significant tricuspid annular dilatation is defined by a diastolic diameter ≥40 mm or >21 mm/m2 measured at end-diastole in the apical four-chamber view [24]. However, measurement of the tricuspid annulus diameter with 2D as compared to 3D echocardiography techniques systematically underestimates the actual tricuspid annulus diameter [25]. Although not routinely used, 3D echocardiography off-line multiplane reconstructions allow accurate measurement of the tricuspid annular dimensions [21].

The approach to assessing TR includes the identification of the type and mechanism of regurgitation, grading of the severity of TR, quantitative assessment of annular dilatation and tethering, assessment of RV dimensions and function, estimation of pulmonary artery pressure and assessment of inferior vena cava (IVC) and suprahepatic vein dimensions [22].

The assessment of type and mechanism of TR requires a careful assessment of leaflet morphology and mobility and annular morphology [22]. Mechanisms of primary TR may involve prolapse of one or more leaflets in degenerative TR, restricted motion of the leaflets as the consequence of rheumatic disease or carcinoid or toxic disease or more rarely leaflet perforation due to endocarditis [22], post-traumatic disease, or congenital disease such as Ebstein’s anomaly.

In functional TR, it is important to determine the relative role of annular dilatation and tethering of the leaflets, as well as to clarify the potential role of additional mechanisms (e.g., interference with pacemaker electrodes).

Significant tricuspid annular dilatation is defined by a diameter ≥40 mm or >21 mm/m2 measured at end-diastole in the apical four-chamber transthoracic view. In functional TR, the degree of tethering of the leaflets can be evaluated by the measurement of the systolic tenting area (area between the tricuspid annulus and the body of the tricuspid leaflets) and the coaptation distance (distance between the tricuspid annular plane and the point of coaptation) in mid-systole from the apical four-chamber view. Severe tethering is defined by a tenting area >1 cm2 and coaptation distance ≥8 mm [21, 22].

Accurate grading of TR remains difficult due to the high dependency of TR to loading conditions, and the lack of objective and reproducible parameters [22, 26] as for mitral regurgitation. Therefore, an integrative multimodality approach including qualitative and quantitative parameters is recommended [22]. Colour Doppler flow provides a qualitative assessment of TR severity. The presence of a large central or eccentric jet reaching the posterior wall of the right atrium is in favour of severe TR. A dense continuous wave Doppler envelope with a triangular contour and early peak velocity, or a laminar TR flow also suggests severe TR [22]. A more quantitative assessment is provided by vena contracta (VC) width measured in the parasternal or the apical four-chamber view. A VC ≥7 mm indicates severe TR, whereas a diameter <6 mm may be either mild or moderate TR. The assessment of proximal isovelocity surface area (PISA) of TR jet by the flow convergence method is not well validated for TR, because of the variable and non-circular morphology of the regurgitant orifice. The published data suggest that an effective regurgitant orifice area (EROA) >40 mm2 or a regurgitant volume >45 ml is indicative of severe TR [22]. A more qualitative approach is to use the PISA radius at a constant aliasing velocity. A TR PISA radius ≥9 mm at a Nyquist limit of 28 cm/s suggests severe TR, whereas a radius <5 mm suggests mild TR [22, 26]. The presence of systolic hepatic flow reversal is specific to severe TR and represents the strongest additional parameter for evaluating the severity of TR [22].

More recently, a new TR classification has been proposed to better characterize the severity of TR currently being treated with various transcatheter devices, including very severe (or massive) and torrential on top of the previous three grades. Currently, no data are available about these subgroups, but this new classification will have a significant impact on the design of future device trials and determine outcomes using quantitative echocardiographic parameters [52] ( Table 2).

Due to the complex shape of the RV, assessment of RV enlargement requires measurement of RV dimensions in multiple views [27]. RV systolic function can be assessed by multiple parameters: fractional area change (FAC), tricuspid annular plane systolic excursion (TAPSE) or tricuspid annulus systolic velocity. Of note, all these parameters have limitations in the presence of severe TR due to their dependence on loading conditions. Despite these limitations, TAPSE <16 mm and tricuspid annulus systolic velocity <11 cm/s could be used to identify patients with RV dysfunction [27, 28]. Systolic pulmonary artery pressure can be estimated from the TR maximal velocity with an estimate of right atrial pressure on the basis of IVC size and collapse with respiration [27]. Estimation of systolic pulmonary artery pressure is no longer possible in the case of laminar TR flow. IVC diameter and respiratory variations should be assessed from a subcostal approach, as well as systolic reversal in the central hepatic vein. The diameter of the IVC is an important parameter for assessing the degree of cardiovascular compensation, as well as for planning heterotopic implantation of transcatheter valves or stent implantation in the IVC. The diameter is measured perpendicularly to the long axis of the IVC at end-expiration, just proximal to the junction of the first hepatic vein which lies approximately 0.5 to 3.0 cm proximal to the ostium of the right atrium [27]. The echocardiographic examination should also look carefully at the potentially associated valvular lesions, particularly on the left side, and assess LV systolic and diastolic function [27, 28].

Intraprocedural echocardiography

Both 2D and 3D TOE play an essential role in addition to fluoroscopy for procedural guidance during transcatheter TV interventions. The availability of 3D TOE, biplane (xPlane) imaging and intracardiac echocardiography (ICE) is particularly helpful during these procedures. In patients with sub-optimal echocardiographic windows, the combined use of transthoracic echocardiography and these advanced imaging techniques may thus be of great utility. Because of the complexity of TV anatomy and the need to improve communication and guidance during intervention, it is crucial that both interventionist and echocardiographer define before the procedure a limited number of imaging views with conventional orientation and anatomical landmarks. Immediate pre-procedural 2D TOE imaging allows comprehensive visualisation of the TV leaflets in the different planes, measurement of TV mediolateral annulus and assessment of TR with colour Doppler ( Figure 2). During the procedure, both 2D and 3D TOE are useful for precise guidance and positioning of devices, as well as for avoidance of complications such as coronary or chamber perforation. A 3D TOE “en face” view of the TV from the right atrial perspective provides an adequate identification of TV leaflets and precise localisation of the anteroseptal and anteroposterior commissures [29] ( Figure 3). Care should be taken to include in the 3D data set the important anatomical landmarks that are the aorta and the interatrial septum [27, 29]. Despite a lower quality of 3D imaging for the TV as compared to those obtained for the mitral valve, a 3D “en face” view is essential for positioning the devices during direct tricuspid valve annuloplasty techniques or for the clipping of the tricuspid leaflet.

xPlane (or biplane) 2D TOE imaging provides simultaneous assessment of two orthogonal views. This mode is complementary to 3D TOE “en face” views and provides accurate identification of the TV leaflet insertion in the annulus, assessment of the course of the right coronary artery in relation to the AV groove, and good overall visualisation of cardiac structures in real time ( Figure 4). Because visualisation of the tricuspid annulus may be difficult with 2D TOE only due to shadowing from prosthetic material, ICE imaging can be used simultaneously with TOE, to overcome this limitation. ICE imaging from the right atrium provides excellent visualisation of the IVC-right atrial junction, tricuspid annulus and relative position of the right coronary artery to the insertion of the TV leaflet ( Figure 4). In addition, the recent 4D volume ICE system (Siemens; Mountain View, California) provides even more accurate images and might find a wide application in this context. Cases of tricuspid clipping under ICE guidance in patients with suboptimal TEE imaging, have been described [60].

Computed tomography (CT)

CT is an essential complementary tool to echocardiography for the pre-procedural planning of several devices. Since it offers a comprehensive assessment of the real 3D anatomy, CT allows the operator to assess the TV anatomy as well as the tricuspid annular shape and dimensions. Multiplanar reconstructions are performed for visualising all three TV leaflets, which are difficult to assess on the same short-axis “en face” view plane because of the non-planar morphology of the TV. The septal leaflet is best seen in the four-chamber view, and a two-chamber view of the right ventricle correctly shows the relationship of the posterior and anterior leaflets. Moreover, the two-chamber view is particularly appropriate for assessing the cavo-tricuspid isthmus superior to the hinge of the posterior TV leaflet, which is more difficult to visualise with TTE. Another major advantage of CT is the ability to combine both an anatomical evaluation of the tricuspid annulus and a semi-quantitative assessment of the tissue characteristics in the right atrioventricular groove. This is based on the measurements of the tissue density (in Hounsfield units) in studied regions of interest (allowing differentiation of calcification, epicardial fat, myocardium, fibrous tissue). This information regarding the structure of the TV annulus is of importance for the appreciation of the tissue quality that supports the leaflets. In case of TV annuloplasty rings, CT is the technique of choice for helping to identify any dislodgement, deformation in the shape, or inappropriate position. Finally, CT is capable of providing data in the adjacent structures including the coronary arteries, the systemic venous drainage and the hepatic veins. In certain pre-procedural evaluations, there is a need to identify the target zone of the implant, to assess the angle between IVC, SVC and the TV plane, to assess the distance between the right coronary artery and the valve annulus and assess the risk of injury of surrounding structures. It is also sometimes necessary to obtain detailed measurements of IVC and SVC dimensions, cavo-tricuspid isthmus length and hepatic veins ( Figure 5).

Certain prerequisites must be fulfilled to obtain technically satisfactory images. Indeed, compared with the mitral valve which can be easily seen in routine cardiac CT, visualisation of the right atrioventricular junction may be difficult and imaging of the tricuspid annulus and leaflets may be challenging. Depiction in detail of this region requires a good and homogeneous enhancement of the structures around the TV annulus, and elimination of any artefact is crucial. The major limitations are the right atrial artefact and superior vena cava artefact, due to the inhomogeneous enhancement of the right atrium and the high-attenuation effect of the contrast medium, respectively. Thus, cardiac CT protocols need modification for the right heart examination, especially with contrast agent administration. These artefacts can be attenuated when injections are performed through the peripheral veins of the lower limb. To obtain optimal images of the IVC, iodinated contrast media is injected through a saphenous vein in the lower limbs with a volume ranging from 80 to 100 ml, at a constant rate of 3 to 4 ml/sec with a power injector, followed by a large bolus chaser of 40-50 ml saline. Acquisitions are performed with retrospective ECG gating, covering a large window including both the heart and the sub-renal IVC or SVC. CT parameters (kV, mA) need to be adapted to the patient’s weight. Good co-operation from the patient is required as data acquisition is performed during a breath-hold period of approximately 8-10 sec (depending on machine characteristics) to avoid respiratory motion artefact. Beta-blockade may be required to lower and stabilise the heart rate, especially if the patient presents with rapid atrial fibrillation in the setting of right heart dilatation.

Current CT machines allow imaging of the heart valves with a high spatial resolution: the minimal slice thickness of 0.5 to 0.625 mm is almost isotropic so that multiplanar oblique reconstruction is obtained without signification degradation of spatial resolution. The tricuspid annulus is assessed on two orthogonal planes passing through the long axis of the RV, and the short-axis plane is obtained as the perpendicular of the two long-axis planes ( Figure 6). Measurements are taken from diastolic phase reconstructions ranging from 60 to 90 percent of the RR interval, during retrospective ECG-gating data sets, selecting the phase with the maximum tricuspid leaflet opening as done with echocardiography.

In the case of annuloplasty devices, pre-procedural imaging of the tricuspid annulus and its environment is performed with the aim of assessing the following goals: 1) delineation of the tricuspid annulus shape and localisation of the three commissures; 2) accurate measurements of the tricuspid annulus diameters and perimeter; 3) localisation of the right coronary artery and its course in the AV groove with particular attention to its relationships and distances from the tricuspid annulus; 4) semi-quantitative evaluation of the soft tissue quality adjacent to the tricuspid annulus.

Current transcatheter replacement opportunities: valve-in-valve and valve-in-ring

As for the mitral position, tricuspid transcatheter implantation of a transcatheter aortic valve (TAVI) prosthesis is a possible therapeutic option for patients with recurrent tricuspid pathology following a surgical implant. Percutaneous valve-in-valve and valve-in-ring implantation has been the subject of numerous case reports, either for a degenerated tricuspid bioprosthesis or for ring annuloplasty failure [30, 31, 32].

The first and crucial step is to evaluate the prosthesis size. The choice of the prosthesis size is derived from an integrated approach, combining the manufacturer’s specifications of inner diameters and the diameter determined by CT, 3D TOE and fluoroscopy imaging. Given the round shape of the prosthesis, a reliable CT scan may be sufficient and accurate. Appropriate device selection depends on the surgical prosthesis, and can be assisted by the valve-in-valve mitral ‘app’. However, due to the relatively smaller experience with tricuspid as compared to mitral valve-in-valve, the sizing strategy has not been completely established. Due to lower closing pressures, oversizing of the TAVI prosthesis is considered to be less crucial in the tricuspid position.

The first implants were done mainly via a transatrial or a transjugular access. More recently, the transfemoral route appears to be feasible and safe, also using local anaesthesia and fluoroscopic guidance or general anaesthesia and TOE guidance. In cases of a narrow caval to annulus angle, a contralateral snare has been successfully used to simplify the implant, by improving coaxiality [33]. After crossing the valve, a J-shaped wire is positioned in the RV apex, and the valve can be slowly deployed under rapid ventricular pacing. Rapid pacing is usually performed by positioning a temporary lead in the left ventricle or in the coronary sinus, unless the patient has a previously implanted permanent pacemaker ( Figure 7). Predilatation of the bioprosthesis is not necessary and it is anyhow not generally recommended due to the risk of leaflet rupture or embolisation.

Two types of valve have been implanted so far in the tricuspid position, the Melody valve (Medtronic, Minneapolis, MN, USA) and the Edwards SAPIEN valve (Edwards Lifesciences) [34]. Both valve devices seem to give good results, but, given the small sizes available for the Melody valve (maximal diameter of 22 mm), it has been utilised mainly in children or young adults with congenital heart disease. Four patients with sizes of the failed bioprosthesis ranging from 25 to 31 mm, have been successfully implanted with Edwards SAPIEN valves, sizes which were not compatible with Melody valves (unpublished data, Himbert et al, Hôpital Bichat, Paris, France).

In the tricuspid position, surgical rings usually have an open configuration, and a non-circular and non-planar shape. As a result, compared to mitral, tricuspid valve-in-ring presents some specific difficulties. Therefore, an integrated approach for the accurate sizing of the ring is mandatory using the three available techniques: CT, 3D TOE and fluoroscopy ( Figure 8). On the other hand, unlike the mitral position, the risk of right ventricular outflow obstruction is absent, and residual perivalvular leak and risk of embolisation are less of an issue as a result of the low-pressure right ventricular system. As a support for prosthesis sizing, during the procedure, balloon sizing may be useful to determine the annulus dimension more precisely [35].

Several reports have been published concerning the Edwards SAPIEN valve, using transatrial and transfemoral approaches [36, 37]. Bouleti et al reported their experience with three valve-in-ring implantations with an Edwards SAPIEN XT valve through a transfemoral approach [38]. The sizes of the rings were between 30 and 32 mm, allowing implantation of three 26 mm Edwards SAPIEN XT valves. On pre-discharge echocardiographic examination, paravalvular leak was absent in one, mild in one and moderate to severe in one. The paravalvular leak was always at the level of the open portion of the ring. At one year, all the patients were alive, with a stable improvement in their functional status (NYHA Class I or II), even in the patient with severe paravalvular regurgitation, and there were no significant gradients (from 3 to 5 mmHg).

Although the ring has the advantage of providing the landmarks and necessary anchoring for a percutaneous valve, it also has the drawback of creating a non-circular landing zone with the inability to seal completely the open segment with the implanted valve ( Figure 8). Therapeutic advances with more conformable or repositionable valves with additional sealing capacity are required to reduce regurgitation after transcatheter valve-in-ring implantation. Therefore, transcatheter valve-in-ring implantation should be considered with caution in these rigid, open prosthetic rings most often used for TV annuloplasty rather than a semi-rigid or flexible one.

Valve-in-native

Coaptation and leaflet devices

TriClip

Edge-to-edge repair with the MitraClip system for mitral regurgitation is a validated technique [44]. This concept has been tested in three patients with severe symptomatic TR deemed at high risk for surgery [45]. The strategy was planned to enable edge-to-edge repair between all three tricuspid commissures requiring three clips using a transjugular approach with a 24 Fr sheath. The procedure was conducted under general anaesthesia and 3D TOE guidance, and also ICE guidance in one patient. The intervention was successfully achieved in all three patients with a mean procedural time of 113.7±42.2 min, without use of contrast dye. There was an acute reduction of TR in all patients (Δ-EROA, 0.4±0.4 cm2), as well as of TV annular diameters (Δ-TV annular diameters, 14.3±8.9 mm). At 30 days, clinical symptoms improved with a measurable effect on NT-pro-BNP levels. More recently, several patients have been treated with a transfemoral approach, and a pilot study is ongoing to assess feasibility and early efficacy of this approach. The feasibility of this approach has also recently been reported [46].

However, the use of the MitraClip system for the TV raised some specific issues due to the complexity of the TV itself, since the MitraClip system was designed specifically for the mitral valve. For this reason, the device has been modified to better adapt to caval-right atrial angle and to TV anatomy (Triclip).

Edge-to-edge therapies has been tested with both MitraClip and TriClip devices and both showed promising results.

According to the TriValve registry including 249 patients treated with MitraClip device, the procedural success achieved 77%, with durable TR improvement in more than two thirds of patients and 19% of mortality at 1-year follow-up. On the other side, the novel TriClip technology was studied in the Triluminate trial: TR grade improvement to moderate or less was observed in 71% of cases at 1-year follow-up. Alongside this, functional status and quality of life improved and a total of 7% of all-cause deaths were reported.

The device demonstrated to be a promising option for the TR management and in the future the combination of edge-to-edge repair technique with another percutaneous technique of reduction of the TR annulus for facilitating the clip placement can also by hypothesized.

Pascal TV Repair System

The Pascal tricuspid valve repair system (TVRS; Edwards Lifesciences) is a 22 Fr system that combines the advantages of leaflet grasping (such as MitraClip) to the physical properties of a spacer (such as FORMA) to overcome some limitations observed with these devices in cases of large coaptation gaps and to further reduce the total regurgitant area. It is currently available also with a narrower conformation without central spacer (Pascal ACE).

Early experience in 28 high-risk patients showed 86% procedural success with 7% of mortality after 30-days follow-up [61]. Data from the CLASP TR early feasibility study confirmed the efficacy and the safety. The device obtained CE approval in 2020.

Mistral

Another promising technology is represented by the Mistral TVRS (Mitralix, Ltd.). It works increasing leaflet coaptation and is made of an 8.5 Fr delivery system and a spiral-shaped repair device creating a “bouquet” of chordae tendinae. Of note, this technology would not preclude the use of annuloplasty or leaflet technology at a later stage. Early results on 7 patients demonstrated safety and efficacy of such innovative device [63].

Annuloplasty devices

Cardioband Annuloplasty System

The Cardioband (Edwards Lifesciences, Irvine, California) is a transcatheter transfemoral direct annuloplasty device that closely reproduces surgical annuloplasty. It is implanted on the atrial side of the tricuspidal annulus from commissure to commissure, with the beating heart under fluoroscopic and echocardiographic guidance.

It is made of three components ( Figure 9):

  • The trans-femoral (TF) Delivery System (DS): the DS consists of the Implant DS (IDS) and the 25 Fr Transseptal Steerable Sheath (TSS). The IDS is comprised of a steerable Guide Catheter (GC) and an Implant Catheter (IC) with the Cardioband implant mounted on its distal end
  • Implantable metal anchors and anchor delivery shaft
  • Size adjustment tool (SAT)

The Cardioband is made of a polyester sleeve with radiopaque markers (ink stamped on the device) spaced 8 mm apart. It is available in six lengths to cover a wide range of annulus circumference sizes comparable to those of commercially available annuloplasty devices. During implantation, 12 to 17 metal anchors are inserted to fixate the Cardioband to the tissue. Each anchor is 6 mm long and is spiral-shaped with a screw-like fixating mechanism designed to secure the implant safely in place. All anchors are repositionable and retrievable until disconnected from the anchor drive. Once the Cardioband is successfully anchored along the annulus, the band is dynamically cinched using a dedicated SAT. The distal tip of the SAT is connected to an implant wire inside the Cardioband. The implant wire is foreshortened by gradually turning on the adjusting knob. Cinching of the band translates into a reduction of the tricuspidal annular diameter of approximately 30%. The effect of reduced annular diameter on improved leaflet coaptation is guided by echocardiography.

Two-years data from the single-arm, multicenter, prospective TRI-REPAIR study were recently published [63].

The study included 30 symptom­atic patients from nine European centres who had chronic severe functional TR in the absence of untreated left-heart disease, tricuspid annulus diameter ≥40 mm, and systolic pulmonary arte­rial pressure ≤60 mm Hg. Technical success was achieved in all patients and only one in-hospital death was recorded. At 2-years follow-up, 8 deaths (27%) occurred. Echocardiographic, functional and symptoms improvements were recorded over a two-years observation period although torrential TR persisted in 9% of patients. Similar results were confirmed also in the early feasibility study [63].

Minimally Invasive Annuloplasty (MIA)

MIA technology (Micro Interventional Devices) is a sutureless annuloplasty system consisting of compliant, self- tensioning low-mass anchors and a thermoplastic elastomer (MyoLast) deployed in the tricuspid annulus. The feasibility and safety study (STTAR trial, not registered) is underway to assess this novel technology among 40 patients with FTR

Orthotopic Replacement devices

Gate Valve ( Figure 10)

The NaviGate (NaviGate Cardiac Structures, Inc.) atrioventricular valve is a biological valved stent; it consists of conical-shaped valve with Nitinol alloy stent into which is mounted a trileaflet valvular mechanism fabricated from pre-treated equine pericardium. The specific design serves the purpose of avoiding flow dynamics alterations and reducing right atrium and right ventricle protrusion. Currently, four sizes are available (40, 44, 48, 52 mm in diameter) and is intended for native tis­sue tricuspid annular diameters of 36 to 52 mm considering that tricuspid annulus for secondary TR is usually 40±5 mm. The valve anchoring is based on specially configured structures, winglets and graspers, 12 located in the proximal part of the stent and 12 in the distal end of the stent. It can be implanted transatrially through a minimally invasive right thoracotomy directly through the free wall of the RA into the TV. Alternatively, the transjugular route has been tested using a 80 cm long flexible and steerable catheter with a 45 Fr (outer diameter) hydrophilic introducer sheath.

So far, data about 30 patients with severe or greater TR have been described: technical success was 87% and the most used access site was the transatrial approach. In hospital mortality was 10%, increasing to 13% at 30-day follow-up. Procedural complications included malpositioning requiring surgical conversion, pacemaker implantation due conduction abnormalities and RV perforation [53].

Evoque

The Evoque system (Edwards Lifesciences) has been tested for compassionate use in 25 patients with at least severe TR. Technical success was achieved in all cases (92%) but two. No deaths were reported at 30-day follow-up, while conduction disturbances requiring pacemaker was needed in 8% of cases [64].

Other Emerging Replacement Technologies

There are currently some new technologies under study in the preclinical phase. The Lux-Valve (Jenscare Biotechnology) is a self-expanding prosthesis made of bovine pericardial tissue onto a nitinol stent frame, implanted through the transatrial approach. Currently, data about the early experience in 12 patients have been published showing promising results [65].

Another innovative technology is represented by the Trisol Valve (TriSol Medical Ltd., Inc., Yokneam, Israel). It consists of a single bovine pericardial piece attached to the nitinol frame and anchoring to the annulus through multiple arms. It is made of two leaflets closing in a dome shaped structure with the aim of avoiding RV sudden afterload increase.

Heterotopic replacement

Caval valve implantation, CAVI ( Figure 11)

Heterotopic TV implantation is an alternative approach to an orthotopic implantation, since, due to the anatomical structure of the TV (dimensions and lack of calcification), solutions for stent and valve fixation has not yet been identified or even reported in human patients with native TR. The novel concept of implanting a dedicated self-expanding valve at the cavoatrial junction has been studied in pre-clinical studies, which demonstrated the haemodynamic effect of caval valves (both implanted in the IVC and SVC) in animals for up to six months [43].

In clinical practice, the goal is to improve the symptoms by protecting the venous system from elevated right atrial pressure and systolic caval backflow in severe TR, without any correction on the TV itself. Two characteristics are needed: the presence of caval backflow

reversal and inferior vena cava diameters ≤30 mm or 42 mm, for balloon-expandable and self- expandable valves, respectively. To date, two types of transcatheter heart valves have been used for CAVI: balloon- expandable valves (29-mm SAPIEN valve; Edwards Lifesciences) [46] and specifically developed self- expanding valves (TricValve, P&F Products Features Vertriebs) and Tricento (NVT) [55, 56].

From a technical perspective, the large and variable diameters of the IVC and SVC and the length of the landing zone between the hepatic veins and inferior cavo– right atrial junction are major challenges

The first-in-man implantation was performed a on compassionate grounds in a 79-year-old woman, with severe functional TR, multiple previous cardiac surgeries and significant comorbidities (logistic EuroSCORE 29.7%) [44]. The prosthetic valve was percutaneously implanted loaded in a custom-made 27 Fr sheath, then delivered at the cavoatrial junction with a safety margin of 5 mm, to avoid a low or high position causing either hepatic vein obstruction or paravalvular leak. At eight weeks, the device was quite stable, there was no paravalvular leakage or remaining venous regurgitation, with improved physical capacity, and the patient was discharged from hospital.

Promising preliminary clinical results have recently been reported in 10 patients treated with the SAPIEN XT valve. A self-expanding stent has been implanted in the IVC to prepare the landing zone for the balloon-expandable prosthesis. Thirty-day clinical and RV longitudinal function improvements were observed, with a significant reduction of the retrograde flow in the IVC [45] (Lauten et al, TVT, Chicago, 2015).

Acute haemodynamic as well as clinical improvements at follow-up have also been reported with a self-expanding prosthesis in five patients, three treated with a dual valve and two with a single valve approach (Lauten et al, San Francisco, TCT 2014).

The first- in-human experience was reported in 2018 and included 25 patients treated with 31 balloon-expandable or self-expandable valves in the caval position [57].

Patients received a single inferior (n=19) or bicaval (n=6) valve implantation, with 92% procedural success, immediate caval backflow reduction and significant functional improvement at 10 months. Two embolizations occurred at 30 days, with a 30-day and 1-year mortality of 12% and 63%, respectively. Therefore, TAVI devices are not used anymore for caval implantation.

There are still some unsolved issues concerning the long-term tolerability of this palliative technique, especially regarding the risk of deterioration of RV function by increasing preload in patients with prior depressed RV function and the right atrium ventricularization. These procedures were performed in patients with preserved RV function. Moreover, if the feasibility and the safety seem acceptable, the procedure is quite expensive using a non-dedicated device, and information on long-term durability are lacking

The TRICUS study (NCT03723239) is currently enrolling and seeks to determine the safety and efficacy of the TricValve device. The TRICAVAL trial (NCT02387697) randomized

patients to optimal medical treatment versus transfemoral implantation of the Sapien XT valve into the IVC. The trial was terminated early for safety concerns after enrolment of 28 patients and the future of this solution to TR is unclear. This technology is also under study in the USA in the HOVER study (ClinicalTrials.gov. Identifier: NCT02339974, Temple University, Philadelphia, PA).

Conclusion

The numerous first-in-man reports recently published and dedicated to the TV, confirm that there is a renewed focus on tricuspid valve treatment, and a clear unmet need for transcatheter tricuspid intervention [47, 48]. The population with severe residual and persistent TR after left-sided heart surgery is growing and often deemed at high risk for subsequent isolated TV surgery. The impact of functional TR on both mortality and morbidity at long-term follow-up requires that we take into careful consideration this ageing population only asking for a better quality of life.

All techniques have specific advantages and challenges; however, it is too early to compare them. They seem feasible and safe, but larger scale clinical studies are needed to confirm these preliminary data. One can also imagine combining these different techniques, for example one reducing the TV annulus and another dedicated to improving leaflet coaptation.

There is an unquestionable prerequisite for being able to achieve success with these innovative techniques, namely, the quality of the imaging before intervention as well as during intervention. Before intervention, data collected using cardiac CT, 2D and 3D TOE imaging are crucial for the selection of appropriate patients. During intervention, 2D and 3D TOE imaging is often used in combination with ICE imaging which itself can give a direct image of the TV. The data collected from TOE and ICE imaging are thus often complementary. Moreover, the ongoing development of multimodality imaging and the possibility of fusing echocardiographic and fluoroscopic images during the intervention could prove to be very helpful in the near future.

Transcatheter tricuspid valve implantation may also become an alternative. However, transcatheter implantation of a bioprosthesis in a native tricuspid valve presents challenging issues, similar to those identified for the mitral valve. Transcatheter implantation of a valve with severe mitral valve calcified stenosis has already been reported, but such an amount of calcification allowing the implanted valve’s position to be secured is rarely present on the TV [48, 49]. Early feasibility of mitral valve implantation on non-calcified leaflets has also been demonstrated. However, the large size of the tricuspid valve and the trileaflet configuration are factors against a direct translation of mitral devices to the tricuspid position.

Personal perspective

Taking into account all the most recent published data for transcatheter treatment of functional TR, new perspectives are awaited in the near future in this pathology which has been neglected for many decades. Using either a transfemoral or a transjugular approach, the access to the tricuspid is easier than of the mitral valve. These new techniques seem feasible and safe in a small number of patients treated in specialised, experienced centres for transcatheter intervention in heart valve diseases. Many issues, however, remain unresolved:

  • How to select patients from among those who are poorly symptomatic with severe TR, very symptomatic (which criteria: dyspnoea only, number of hospitalisations, amount of diuretics) and what to do in patients with end-stage right heart failure.
  • The population with functional TR is heterogeneous. Those with isolated TR (most often an elderly population with a long history of permanent AF) are probably not comparable with those with previous left-sided heart surgery and uncorrected TR at the time of surgery.
  • Timing: the best timing for intervention needs to be established, yet. Current data demonstrate how late are transcatheter interventions in most patients enrolled. Fairly often Heart Team assessment finds TV interventions either too early or too late. A deeper knowledge of TR progression and a careful imaging characterization might help to define the best timing for intervention
  • Which anatomical criteria are the most appropriate for guiding the decisions? TV annulus diameter or severity of the TR, taking into account that echocardiographic assessment is both non-reproducible and unreliable.
  • What are the best endpoints? Clinical outcome only and/or echocardiographic parameters (reduction of TV annulus or the severity of the TR).

There is still a lot of work in prospect for the selection of the “ideal” candidates and the “accurate” evaluation criteria. What is the best transcatheter technique? We have to consolidate the feasibility and safety data before the widespread adoption of any technique. The final step will be to assess the efficacy both in the short and in the long term. Studies with larger numbers of patients and longer follow-up time are, however, still necessary.

SHARE YOUR COMMENT